Exploit Development

Authorization Boundary

• Work only on owned code, lab targets, CTFs, or explicitly authorized research.
• Keep proof of concept minimal: demonstrate the bug and impact without persistence, evasion, automated exploitation at scale, or post-exploitation.
• Prefer root-cause analysis, exploitability classification, mitigations, and regression tests.

Workflow

1. Establish target version, build flags, architecture, mitigations, input vector, and crash artifact.
2. Reproduce deterministically in an isolated lab with symbols and sanitizer output when possible.
3. Triage root cause: bounds, lifetime, type confusion, race, injection, logic flaw, or unsafe parser behavior.
4. Assess exploitability at a high level: control of instruction pointer, write primitive, info leak, sandbox, and mitigations.
5. Provide a safe PoC or pseudocode only as needed to validate impact, then produce patch guidance and tests.

Defensive Outputs